The effects of seat-belt regulations on automobile-related fatality and injury rates have been of great interest to economists and policy-makers over the past few years.' The effects of the laws have been evaluated by various statistical techniques using timeseries data for particular states and pooled time-series data for national models.2 The results of these studies provide some evidence that seat-belt laws (SBL) reduce injury and fatality rates. However, the effects of seat-belt laws vary across states and time periods as well as across the levels of injuries. This study assesses the effects of the Texas seat-belt law on injury rates using policereported accident data. The data are from the U.S. Department of Transportation State Traffic Accident Files and are compiled monthly for the period 1982-1987 for driver-involved accidents. Furthermore, the data comprise singleand multiple-vehicle accidents. Only accidents involving towed vehicles are used in the analysis so as to normalize for changes in accident-reporting thresholds over time.3 The analysis was conducted for several sets of injury classifications using the KABCO scale, which indicates the numbers of fatalities (K), severe injuries (A), moderate injuries (B), complaints of injuries (C), and no injuries (0).

I use micro data from the Labor Force Surveys conducted in the West Bank and the Gaza Strip during 1981-1991 to show that during 1981-1987 wage differences between schooling groups fell by well over one-half. This sharp reduction is associated with large increases in the size of the educated Palestinian labor force. Since the returns to schooling for Israeli Jews were stable, the decline in returns to schooling for Palestinians is consistent with the notion that the returns to schooling in the territories were determined largely by the forces of supply and demand in a segmented market for skilled labor.

We investigate the structure of preferences and uncertainty that guarantees that prices are fully revealing even though asset markets are incomplete and there are more sources of uncertainty than assets in the economy. A sufficient condition for fully revealing prices is that investors have preferences of the (possibly state-dependent) linear-risk-tolerance class. Finally, we discuss how our result allows one to extend certain existing literature on demand aggregation, welfare analysis, and the pricing of contingent claims to the case in which markets are incomplete and investors have asymmetric private information.

The frequency of foreign conflict initiations in the United States is found to be significantly greater following the onset of recessions during a president's first term than in other periods. The authors develop an economic theory of the political use of wars which links the election cycle, war decisions, and economic performance consistent with the observed relationships among these events. An incumbent leader with an unfavorable economic performance record may initiate a war to force the learning of his war leadership abilities and thus salvage, with some probability, his reelection. This obtains despite voter rationality and informational symmetry.

The frequency of foreign conflict initiations in the United States is found to be significantly greater following the onset of recessions during a president's first term than in other periods. We develop an economic theory of the political use of wars which links the election cycle, war decisions, and economic performance consistent with the observed relationships among these events. An incumbent leader with an unfavorable economic performance record may initiate a war to force the learning of his war leadership abilities and thus salvage, with some probability, his reelection. This obtains despite voter rationality and informational symmetry.

We review and provide motivation for a one-sector model of economic growth in which decisions about capital accumulation are made by a political process. If it is possible to commit for at least three periods into the future, then for any feasible consumption plan, there is a perturbation that is majority-preferred to it. Furthermore, plans that minimize the maximum vote that can be obtained against them yield a political business cycle. If it is impossible to commit, voters select the optimal consumption plan for the median voter.